The long term goal of this project is the elucidation of the mechanism of pathogenicity of the pneumococcus in ocular infections. Our previous efforts have focused on a toxic protein, pneumolysin, which is produced by all pneumococci. This protein is a cytolyic agent, attacking all cells containing cholesterol in their membranes. Earlier studies focused on purification of die agent, study of its properties and interactions with PMNs and on its basic mechanism of action. More recently we have cloned the gene encoding pneumolysin and constructed strains of pneumococcus in which the gene is deleted. Comparisons of the virulence of such strains with the wild type in a rabbit model of intracorneal infection have shown greatly reduced virulence in the deletion mutant, confirming our hypothesis that pneumolysin plays an important role in the pathogenesis of ocular infections. Other studies in this laboratory and other laboratories suggest that one mechanism by of complement, a function involving a with cytolysis. The studies we propose in the present application will: 1) Confirm the role of pneumolysin in this model by moving the gene back into the strain bearing the deletion, thus restoring the wild type phenotype, to see if virulence is restored to wild type level; 2) Determine whether the underlying mechanism involves complement activation rather than the cytolytic function by assessing the virulence of pneumococci bearing a gene coding for a mutant pneumolysin which retains cytolytic function but does not have the ability to activate complement. In addition, we wish to study another potential virulence factor, pneumococcal neuraminidase, which may play a role in adherence of the pneumococcus to ocular surfaces and in the early stages of infection. We will perform basic studies on neuraminidase including a survey of its incidence and isozyme number in strains of various origin, and purify the enzyme(s). We will clone the gene, and construct deletions as we have done for pneumolysin, and then compare the deleted mutants and wildtype strains in systems we will develop for the study of the role of neuraminidase in adherence and early infection. The latter will involve in vitro studies of corneal epithelial cells and in vivo studies of animals with corneal epithelial damage. We will also assess the effect of prior induction of mucosal immunity to neuraminidase on response to pneumococcal challenge in these models. We expect that these studies will add to our knowledge of the pathogenesis of the pneumococcus in the eye, but the results may also be relevant to other ocular infections and to pneumococcal infections in other parts of the body .